Studies in our laboratory have demonstrated that macrophages can directly affect erythropoiesis, both in vitro and in vivo. Using a unique model system developed in our laboratory, we have found that this macrophage regulation of erythropoiesis is impaired in leukemia. Regulation can be restored and the leukemia reversed by the treatment of fully leukemic animals with normal macrophages. When macrophages are transferred to normal or leukemic animals, erythroid colony formation (CFU-E) is significantly suppressed. The data suggest that macrophages do not suppress erythropoiesis in vivo through regulation of EPO levels but by an influence on novel erythropoietic stimulatory factor. This factor is produced in response to anemic stress and is present in impure but not purified preparations of EPO. The factor can be measured by its ability to reverse macrophage suppression of CFU-E and has no activity in the traditional in vivo standard plasma EPO assay. Preliminary characterization reveals that the activity is a glycoprotein of approximately 40,000 daltons. No suppressive activity has been detected by normal macrophages in vitro. The production in vitro of a CFU-E stimulatory factor by virus-infected macrophages correlates with the failure of leukemic animals to experience regression. Our goals for the coming year are to: (1) extend our present studies to include isolation and characterization of the factor related to EPO that appears to be regulated either directly or indirectly by normal macrophages; (2) determine the factor's role in vivo in regulation of erythropoiesis; and (3) characterize the in vitro stimulatory factor produced by virus-infected macrophages. (MB)